We propose to conduct a historical evaluation of worker exposures under OSHA's 1984 ethylene oxide (EtO) standard. Most potentially-exposed workers are women employed in the hospital sector where EtO is used to sterilize medical devices. Quantitative and qualitative intervention research methods will be applied to three existing sources of data: the Massachusetts Hospital EtO Health and Safety Study, OSHA's Integrated Management Information System (IMIS), and the nationwide database of Kern Medical Corporation, a commercial vendor of EtO passive dosimeters. Our specific aims are to: 1) Characterize EtO Exposures: Characterization will include summary measures of EtO exposures over time in the hospital and other industries and preliminary characterization of exposure distributions by industry, by workplace, and by individual worker. To the extent feasible, job title- or job task-based characterizations will also be conducted. 2) Identify Potential Determinants of EtO Overexposures: Multivariate regression models of the potential determinants of the three types of EtO overexposure events (Yes/No) and two types of quantitative exposure levels (average workshift and short-term) will be developed using data from the Massachusetts Hospital Study and the Kern Medical Database. Independent variables include engineering controls, sterilizer types, EtO usage frequency, monitoring frequency, and EtO health and safety training. This quantitative approach will be complemented by qualitative assessments of the relationships between the passage of the EtO standard, the implementation of control measures, and changes in exposure levels over time. 3) Assess Compliance with the EtO Standard: To assess the extent to which OSHA enforcement efforts have detected non-compliance, we will compare OSHA IMIS database estimates of the prevalence of non-compliance with the EtO standard with independent estimates from the Massachusetts Hospital Study and the Kem Medical database. These analyses will be conducted for compliance with all principal sections of the standard as well as with exposure limits. Finally, using the Kem Medical database alone, judgments of compliance with OSHA's EtO exposure limits will be compared using traditional OSHA approaches (exceeded limits or not) versus an alternative approach that incorporates all collected exposure data into a measure of the likelihood that true mean exposure exceeds limits.